observability

# Health check
lettactl health [-q] [-v]

# Context usage
lettactl context <agent> [-o table|json|yaml]

# Async runs
lettactl runs <agent> [--limit <n>] [--status pending|running|completed|failed] [-o table|json]

ContextUsage {
  total_tokens: number
  max_tokens: number
  usage_percent: number
  breakdown: { system_prompt, memory_blocks, tool_definitions, conversation: number }
}

Run {
  id: string
  status: 'pending' | 'running' | 'completed' | 'failed' | 'cancelled'
  created_at: string
  completed_at?: string
}

# Check server
lettactl health

# View context usage
lettactl context my-agent

# Get usage as JSON
lettactl context my-agent -o json

# List recent runs
lettactl runs my-agent

# Show only failed runs
lettactl runs my-agent --status failed

# Show last 50 runs
lettactl runs my-agent --limit 50